A public notebook for what I learn while building, testing infrastructure, running benchmarks, and deciding which tools are worth keeping.
What worked, what broke, and what changed my mind while building with AI.
Local runtimes, Cloudflare, containers, deployment paths, and operational tradeoffs.
Practical notes from testing models, agents, evals, and tools.
Small utilities, extensions, automations, and experiments that become useful enough to document.
Posts are grouped by theme so you can find an idea again without scrolling a feed.
A decision memo on moving from agent-by-agent coordination to Firstmate and Herdr: isolated worktrees, durable state, supervised loops, decision gates, and evidence-backed PR handoffs.
A decision memo on moving from agent-by-agent coordination to Firstmate and Herdr: isolated worktrees, durable state, supervised loops, decision gates, and evidence-backed PR handoffs.
The Autonomy alpha is live — a native macOS app and local MCP server that lets AI agents work through your real accessibility setup — plus the day-three build-log of shipping it with Claude Fable 5.
I thought I had built Accept-header Markdown for AppHeat. Production had other plans, because Cloudflare was skipping the Worker entirely.
A field note from setting up a local Pi workflow with OpenRouter agents to research AEO topical authority. The real surprise was not any agent tactic — it was what Google's own guidance actually says.
A field note on handing a large Rust refactor to Fable: knowledge-graph analysis, a full graph rebuild that survived a mid-run failure, and a 9,859-line file cut to 877 across six verified commits.
A practical note on making pages easy to read, cite, and hard to misquote.
A field note from adding memory to a Bible study agent app: start with user-visible records, then add semantic retrieval and traces.
A practical benchmark comparing Apple's container runtime and Colima across Redis, Postgres, ClickHouse, DuckDB, and Neo4j on one M4 Mac.
